MSER Based Text Localization for Multi-language Using Double-Threshold Scheme
Journal Title: EAI Endorsed Transactions on e-Learning - Year 2015, Vol 2, Issue 7
Abstract
In this paper, a region-based text localization that is robust for multiple languages is presented. Maximally Stable Extremal Regions (MSERs) are used for detecting candidates of text areas. The MSER components are grouped based on their connectivity in a feature space by using a new proposed rule for assigning the connectivity. The groups of components are classified into three classes that are text regions with high confidence, text region with low confidence, and non-text regions. A chain of text attribute constraint decision with the double-threshold scheme is developed to identify text regions. A sequence of constraint decision is designed to minimize the complexity based on short-circuit evaluation of logic operators. The regions that satisfy all strong constraints will be considered as text regions with high confidence while the regions that fail in some strong constraints but satisfy all weak constraints will be considered as text regions with low confidence. The final text regions are obtained from all text regions with high confidence and text regions with low confidence that have connectivity to text regions with high confidence. The proposed scheme is evaluated by using the natural scene images that consist of totally nine languages with different text alignments and camera views. The experiment shows that our proposed scheme can provide the satisfy results in comparison with baseline method.
Authors and Affiliations
Chayut Wiwatcharakoses, Karn Patanukhom
Keywords
Related Articles
- EP ID EP45950
- DOI http://dx.doi.org/10.4108/icst.iniscom.2015.258413
- Views 250
- Downloads 0
Effects of using learners’ produced screencast as worked examples in learning
In this study, we compared the effect of learning by worked example and the cognitive load imposed by learner creating or using screencast in three conditions; studying worked examples (USER), example-problem pairs (PRUS...
Keep me posted! Human and machine learning analysis of Facebook updates
The key element of Facebook social network platform is the status updates, in which the user can upload text or other media such as pictures and videos. In this study, we manually classified more than 3500 Facebook statu...
ThesesDB – blended self-service and supervision of students’ theses
e-learning can be seen as service creation process between students and supervisors. In order to automate and enhance thesis and seminar processes at our university, we aim to increase transparency, traceability, communi...
KPIs identification for evaluating E-learning courses through students’ perception
The use of e-learning in education is an ever-increasing practice. E-learning could generate effective learning for education. There are several factors affecting the creation of successful e-learning for education as we...
The antecedents of e-learning adoption within Italian corporate universities: a comparative case study
The implementation of Information and Communication Technologies (ICT) in business education appears to be influenced by a number of organizational issues, such as culture and technological sophistication. However, extan...